from __future__ import annotations

import os

import numpy as np
import torch
import torch.nn as nn

from mmdet.apis import inference_detector, init_detector
from mmpose.apis import inference_top_down_pose_model, init_pose_model, process_mmdet_results

#os.environ["PYOPENGL_PLATFORM"] = "egl"

# project root directory
ROOT_DIR = "./"
VIT_DIR = os.path.join(ROOT_DIR, "vendor/ViTPose")

class ViTPoseModel(object):
    MODEL_DICT = {
        'ViTPose+-G (multi-task train, COCO)': {
            'config': f'{VIT_DIR}/configs/wholebody/2d_kpt_sview_rgb_img/topdown_heatmap/coco-wholebody/ViTPose_huge_wholebody_256x192.py',
            'model': f'{ROOT_DIR}/_DATA/vitpose_ckpts/vitpose+_huge/wholebody.pth',
        },
    }

    def __init__(self, device: str | torch.device):
        self.device = torch.device(device)
        self.model_name = 'ViTPose+-G (multi-task train, COCO)'
        self.model = self._load_model(self.model_name)

    def _load_all_models_once(self) -> None:
        for name in self.MODEL_DICT:
            self._load_model(name)

    def _load_model(self, name: str) -> nn.Module:
        dic = self.MODEL_DICT[name]
        ckpt_path = dic['model']
        model = init_pose_model(dic['config'], ckpt_path, device=self.device)
        return model

    def set_model(self, name: str) -> None:
        if name == self.model_name:
            return
        self.model_name = name
        self.model = self._load_model(name)

    def predict_pose(
            self,
            image: np.ndarray,
            det_results: list[np.ndarray],
            box_score_threshold: float = 0.5) -> list[dict[str, np.ndarray]]:
        image = image[:, :, ::-1]  # RGB -> BGR
        person_results = process_mmdet_results(det_results, 1)
        out, _ = inference_top_down_pose_model(self.model,
                                               image,
                                               person_results=person_results,
                                               bbox_thr=box_score_threshold,
                                               format='xyxy')
        return out